Autoguiding Issue

[Likwid]

Hello, I am getting very frustrated with my setup and I am completely lost. So here is my rig: Rebel T3i Modded on an Orion ED80 for imaging, ZWO ASI120MM on an Orion ST80 for Autoguiding, all on an Orion Siruis Mount. My rig weight is about 15 lbs, which is half the max payload of the Orion Sirius. I spent a lot of time balancing and calibrating everything and I have not had any problems until about 2 weeks ago. I splurged and bought an Astronomik Ha 12mm Clip-In filter for my T3i. I have had no problems getting 5 minute frames consistently, so I figured it was time to upgrade a little bit. That night, something changed and I could not get more than 60 seconds out of anything. I gave up and figured it was some random issue. The next clear night, I had the same thing happen. I realized that my guide scope might be slightly out of focus, so I got my Bahtinov mask out and made sure it was dead on. I pushed my rig and got a 7 minute exposure that night, it was great. By the time I had finished screwing around with everything, it was time to go in. Fast forward to tonight, I thought I had everything figured out. I got the bahtinov mask on both my guide scope and imaging scope, everything was perfectly in focus and I tried a 60 second test exposure. Now, I started out with a Star Adventurer so I am really finicky about my alignment. I always level my mount, assemble it and do a 2 star alignment. Tonight, my 60 second exposure looked like I had a 600mm lens on a normal tripod, it was horrible. After I got all my gear inside, I thought it could be a balance issue, so I rebalanced everything. I went back out, realigned, and same thing. I am at a complete loss and would appreciate any help. My next step is probably going to be completely disassembling my mount and giving everything a good cleaning and lube, to make sure no gears or anything are damaged. The odd thing is that my PHD graph looks great, but the software almost loses sight of my target star. If you look at my screenshot, the square and centered star are not on the crosshair. You can see my horrible 120 second exposure of M13 in the background. Help! Thank you in advance!

[PeterCPC]

Do you run a set of darks in PHD before you start guiding? I am wondering if you are trying to guide on a hot pixel.

Peter

[Laurin Dave]

Looks to me as if it’s only guiding in RA, which is all the StarAdventurer can do. I’d check that you have the right settings/profile selected in PHD.. the bottom right box on the PHD image says “off”

Edited September 10, 2019 by Laurin Dave

[Stu Wilson]

I had the guiding on hot pixel issue for a few runs last year before I figured it out.

Mines an Orion autoguider and a darks library sorted this out.

[Hughsie]

2 hours ago, Laurin Dave said:

Looks to me as if it’s only guiding in RA, which is all the StarAdventurer can do. I’d check that you have the right settings/profile selected in PHD.. the bottom right box on the PHD image says “off”

As Laurin mentions, click on the brain icon and select the ‘Algorithms’ tab. Look at the bottom right hand button under Dec. You may have “Dec Guide Mode” switched to off in which case turn it on.

Then take some dark frames, calibrate and run the guiding assistant from the tools menu for at least two minutes and apply any suggestions it makes.

J

 

Edited September 10, 2019 by Hughsie

[ollypenrice]

According to Craig Stark, creator of PHD, soft focus on the guide camera is actually a slight advantage since it helps in the calculation of the position of the centroid so it won't be that. Mounts don't need to be level so forget that as well. Like the others I think this will either be a setting issue (guiding only on RA looks promising) or a cable/poor contact problem. It's as well to have a full set of spare cables anyway so I'd buy them and try them. If they are not the culprit they will be one day!

Olly

[Likwid]

Oh wow, thank you everyone for the quick answers! I will definitely try new cables and look at my settings. It never even occurred to me that I could be trying to guide on a hot pixel, so I will definitely make use of the darks. There are so many things that can go wrong and everything is so complicated that I don't even know where to start troubleshooting after the basics. It looks like it will be about a week or so until the next night of clear skies, so I have plenty to do until then. Thank you all again, it is much appreciated!

[cjdawson]

One more thing that you might want to try is extending the exposure time from 1 second to 2 seconds.  My experience is that longer exposures help the with guiding as well, as having too short an exposure means that you could end up "chasing the seeing"    In case you don't know, that's the wobble that happens in the atmosphere caused by air turbulance.   The more wobbly that the seeing is, the more that the guider would try to correct when using short exposures (1 second is about the minimum that I'd ever attempt).

[Anthonyexmouth]

+1 for being too in focus on your guide, a little off focus can help. 

[Freddie]

How is your PA? You make no mention of how you do it or how good it is.

[Likwid]

So I just set everything up quick in my garage to check PHD Guiding and sure enough, DEC guiding was off. I have no idea how that changed. I feel like an idiot, but I would have never thought to check that!

Laurin Dave: I guess that was bad wording. I am using an Orion Sirius, but one of the habits that I got from using my Star Adventurer previously was always making sure my mount is level and getting my alignment spot on.

Freddie: I always level my mount even though I know I don't need to. I am using the Hand Controller currently for everything. I built a cable for EQMOD, but I am waiting to add that into the mix until I get more comfortable with everything else. The hand controller tells me where Polaris need to be, I double check it with an app that I have on my phone and make sure that it is dead center in the circle after my mount is perfectly level. I never take shortcuts with my polar alignment. I'm not sure if that is a good or bad thing

Cjdawson & Anthonyexmouth: That is exactly the opposite of how my brain was comprehending autoguiding. I figured if I had longer exposures, the mount would be going longer without correcting and the guiding wouldn't be as good. I set it to 1 second so it could update the view and correct sooner. Now that I think about it though, if I am able to take like 60-90 second exposures unguided, then 2 seconds won't make a huge difference.

Thank you again everyone, hopefully all I needed to change was the DEC guiding. I will never know how that even changed, very odd. I did learn a lot of other things to look for, so hopefully that will make things much smoother in the future. Thank you all again. I don't post very often but when I do, I always get a lot of support. It makes the process of trying to wrap my head around this incredibly complicated hobby much less painful.

[Anthonyexmouth]

3 minutes ago, Likwid said:

 

Freddie: I always level my mount even though I know I don't need to. I am using the Hand Controller currently for everything. I built a cable for EQMOD, but I am waiting to add that into the mix until I get more comfortable with everything else. The hand controller tells me where Polaris need to be, I double check it with an app that I have on my phone and make sure that it is dead center in the circle after my mount is perfectly level. I never take shortcuts with my polar alignment. I'm not sure if that is a good or bad thing

 

It's worth looking into using sharpcap for PA or drift align with PHD2.  I can't have looked  through my polar scope more than a dozen times. saves the knees and its more accurate. 

[alacant]

4 hours ago, Hughsie said:

“Dec Guide Mode” switched to off in which case turn it on

Hi

IIRC, the OP's mount can only guide in RA, DEC should be set to off and no algorithm should be selected.

If all else fails, you could try the rather exellent PHD2.

HTH

[Likwid]

50 minutes ago, alacant said:

Hi

IIRC, the OP's mount can only guide in RA, DEC should be set to off and no algorithm should be selected.

If all else fails, you could try the rather exellent PHD2.

HTH

Sorry, I had poor wording in my first post. I am using an Orion Sirius. I was trying to say that because my first mount was a Star Adventurer, I picked up habits like making sure my mount is perfectly level when I know I don't need to do that with the Orion Sirius. I also didn't even know about PHD2, I am going to give that a shot as well. Thanks!

[alacant]

2 hours ago, Likwid said:

poor wording

OTC, my fault for not reading your post properly.

So yeah, you're gonna need to guide both RA and DEC to avoid the trailing.

Cheers.

[blinky]

and 2 second images are better than 1, as had been said, you will be 'chasing the seeing' in 1 second guide frames

[spillage]

I find for me that 2.5 seconds in phd works well in most conditions but be prepared to make small adjustments as needed and I guess its very dependant on the kit you use. Also ditch the handset, bite the bullet and use your eqmod lead.  You will find it so much easier and you will get rid of allot of mucking about with the handset settings.

I have never used backyard eos only apt and sgpro but with the eqmod lead you just start up you software connect the gear and platesolve to you target..job done. I just kick myself wondering why I took so long to do it this way but I guess like you I thought I needed to build up to it..

[Likwid]

That actually makes a lot of sense now that I think about it. I'll definitely start using longer exposures with guiding. I also spent a large chunk of my day reading about platesolving and alignment in EQMOD and it doesn't really seem as bad as I thought it would be. I think next session I am going to start with my normal workflow just to make sure that my guiding issue is fixed and then I'll take the plunge and start figuring out EQMOD. I guess I have a bad case of "I don't know what I don't know". I'll just have to keep posting and learning! Thank you again everyone!

[spillage]

Check out Chris Shillito on youtube here he is one of the developers of eqmod. He is also a member of this site.

[Likwid]

Oh, perfect! I followed him on YouTube. Hopefully the videos will make the learning curve easier. Thanks!

[Likwid]

So I am currently in my backyard and I am still having issues. PHD1 had the same problem after enabling DEC so I installed PHD2 and built a dark library. When PHD2 was doing the initial calibration, it said that it detected little south movement ad to check the mechanics of my mount. I moved to 2 second exposures on PHD2 also. I just did a 120 second exposure and it looked great, so I moved to 240 and it was perfect! The new graphs don't look very good so far, and it is information overload right now lol. I am going to see how far I can push my exposures and then I am going to get EQMOD all set up and go from there. Thanks again everyone for all the help!

Edited September 13, 2019 by Likwid
Hit submit too early

[alacant]

5 hours ago, Likwid said:

little south movement

Hi

That message is simply FYI. It has no bearing on the guiding which takes only the north mount movement into account.

Don't worry that the graphs look bad; that's the next stage. But in any case, you are quite correct to look at the images taken with your imaging camera. It is all too easy to become a graph watcher!

Congratulations anyway:)

[ollypenrice]

On 10/09/2019 at 14:58, cjdawson said:

One more thing that you might want to try is extending the exposure time from 1 second to 2 seconds.  My experience is that longer exposures help the with guiding as well, as having too short an exposure means that you could end up "chasing the seeing"    In case you don't know, that's the wobble that happens in the atmosphere caused by air turbulance.   The more wobbly that the seeing is, the more that the guider would try to correct when using short exposures (1 second is about the minimum that I'd ever attempt).

 

On 10/09/2019 at 22:33, blinky said:

and 2 second images are better than 1, as had been said, you will be 'chasing the seeing' in 1 second guide frames

I haven't done any experimenting on this matter but recent thinking challenges the 'chasing the seeing' model. The discussion soon gets quite involved but here it is: https://www.cloudynights.com/topic/652092-autoguiding-simulator-benefits-of-short-guide-exposure/  A key idea is that, in using short exposures, it's important not to let corrections 'overlap.' It is this which will certainly cause instability.

While I use 3 to 4 second exposures on our Mesus I've had good results using 0.5 seconds on our EQ sixes. The problem is, though, that a flat graph from short guide exposures could be 'flat' only in respect to the perceived position of the star which might be seeing-affected. How could we test, objectively, the difference between short and long guide subs? One way would be to do two runs, guiding first in longer subs then in shorter. On both runs we would set the camera to take, but not download, images of a long enough duration to be fairly seeing-unaffected. Say 30 seconds. We'd have the capture software reading the FWHM of one star on each of these images and we'd write it down. Once we had a good set of FWHMs under both guiding régimes we could average each of them and see which régime gave the lowest FWHM.

It would be interesting to have some hard evidence on the matter of guide sub length because the trace itself doesn't provide it since it doesn't know where the star 'really is.'

Olly

Edited September 13, 2019 by ollypenrice

[blinky]

That's an interesting thought, will go read the article.  

[cjdawson]

I'll have to read that article in full as well.   I've had a quick scan through some of the top of the thread, but there was not much talk about atmospheric shift.  Which is what the "chasing the seeing" thing is all about.  I do agree that if your mount is setup perfectly and fully calibrated then it should not need to make many guide corrections.  However, making a correction is in my opinion correcting for an error that has already occurred.  So with that thought, you can make as many corrections as you like as it will reduce the overall error in the image, but there will be error as a correction was needed.      The real question is how much error is acceptable for you?

The things that they are talking about in the simulator look to me like things that are mechanical and to do with the mount or optical train.  However, the seeing isn't something that we can control hence why it's better in reality to make sure that your exposures on the guide camera are long enough to make the seeing fall into the background noise of the image, rather than a motion that needs to be corrected.